Vincristine is active against a wide variety of malignancies and has been shown to substantially improve outcomes. Although vincristine is among the most commonly used anticancer agents, little is known about optimal therapeutic dosing and it is widely recognized that improper dosing can lead to serious side effects or lack of efficacy. Vincristine is associated with highly variable neurotoxicity that often necessitates dose reductions, thereby compromising efficacy. Recently published data indicate that vincristine pharmacokinetics may be associated with long-term outcomes in children with acute lymphoblastic leukemia (ALL). Two enzymes (CYP3A4 and 3A5) metabolize vincristine; but CYP3A5 is 10-times more efficient as a catalyst of vincristine metabolism. Severity of neurotoxicity may be directly related to an individual patient's vincristine exposure. It may be possible to optimize vincristine dosing based on knowledge of genetic polymorphisms responsible for vincristine metabolism that alter drug exposure and the risk of neurotoxicity. The long-range goal of this research is to optimize the use of this critically important drug. The objective of this proposal is to assess whether candidate gene polymorphisms or other less obvious genetic variants identified using a genome wide approach are predictive of vincristine neurotoxicity. The central hypothesis is that germline genetic polymorphisms are associated with vincristine toxicity, pharmacokinetics, and efficacy. The first and second aims are to evaluate the association of pharmacogenetic polymorphisms in candidate genes and in the vinca alkaloid pharmacologic pathway, respectively, with vincristine toxicity, efficacy, and pharmacokinetics. Given that other genes may be involved and genome-wide approaches may miss important genetic associations, we propose a candidate pathway approach as aim 2. These aims will involve enrollment of a single cohort of 175 children with precursor B cell ALL to a multicenter prospective clinical trial. DNA and pharmacokinetics will be collected and patients will be followed throughout their treatment for evidence of vincristine neurotoxicity using the standard NCI Common Terminology Criteria as well as more specific and sensitive neuropathy assessment tools (validated in adults) that we will attempt to validate in children. Aim 3 is to test for the association of multiple genetic polymorphisms with vincristine toxicity using a genome wide approach in children with ALL enrolled to completed cooperative group trials in which whole genome arrays have been completed. A second phenotyped population (175 children enrolled on multicenter trial above and 140 children enrolled to a nearly completed trial of pharmacogenetics of vincristine neurotoxicity) will be used as a validation cohort. We expect that this research will provide important new information regarding the association of genetic variables and vincristine toxicity, efficacy, and pharmacokinetics. The results will be significant because they address an important gap in knowledge and will provide the basis for subsequent studies aimed at optimization of vincristine dosing for individual patients in treating curable pediatric diseases.